S. Talha Bukhari

Inconsistency in Jaccard Loss Implementation

1

I have come across a few implementations of Jaccard Loss that vary a lot. Among all those, the one by [DeepMedic](https://github.com/deepmedic/deepmedic/blob/master/deepmedic/neuralnet/cost_functions.py#L31) made more intuitive sense to me, as the 'per-class'...

#526 **- What I did** Fixed the issue referenced, which concerns implementation of Jaccard Loss function. **- How I did it** I changed the **axis** argument of summation terms from...

Classification Layer - Activation & Shape?

2

Hi! I went through this repository after reading your [paper](https://arxiv.org/abs/1705.03820). Architecture on page 6, shows the final classification layer to produce feature maps of shape (240, 240, 2) which may...

Unexpected `shape` issue in Hessian-Vector computation

4

Hi! Thank you making the source code of your work available. I tried to use the library for an application involving a 3D network architecture, and ran into the following...

Question: No New Net (2019)

2

Hi! My question is about your paper [No New Net](https://arxiv.org/abs/1809.10483) which cites your previous [paper](https://arxiv.org/abs/1802.10508), corresponding to this repository. Are you using **Concatenation** to supply feature maps from encoder pathway...

Unused variable: `point_vels`

1

Hello, The `point_vels` variable is not used in loss computation: https://github.com/mikh3x4/nerf-navigation/blob/7fd14ff6640602e49e92e2a54d40ba0e519debae/nav/quad_plot.py#L232 I am facing issues in the initial path planning and I believe this is relevant.